Epilepsy is one of the most common neurological disorders. Yet, very little is known about how seizures start, spread and terminate. Progress thus far has been hampered by the challenge of monitoring the activity of ensembles of single neurons in humans. Most studies have been limited to intracranial electroencephalograms (iEEGs). Animal models have been used as an alternative approach, but it remains an open question how well these animal models capture mechanisms underlying human epilepsy. Recent technological advances will allow the present proposal to meet this challenge, through the simultaneous recording of large ensembles of single neurons in humans with focal epilepsy, during interictal, preictal, ictal and postictal periods (Truccolo et al., 2011). Patients with pharmacologically intractable focal epilepsy will be recorded during pre-resection surgery monitoring. This will be accomplished using intracortical 96-microelectrode arrays (96-MEA, 4 mm X 4 mm), in addition to subdural iEEGs. The activity of large neocortical ensembles of single units (SUs), multiunits (MUs) and high-spatial resolution local field potentials (LFPs) will be recorded continuously (24hr/day over a period of ~1-2 weeks). Three specific AIMs will address three main fundamental and inter-related problems in the neurophysiology of human focal epilepsy. AIM 1 will determine the role of interictal discharges (IIDs) in seizure facilitation/inhibition and relate the activation patterns of neuronal ensembles during IIDs and ictal periods. Some previous studies propose that IIDs initiate a condition that preludes the onset of an ictal event. By contrast, other work hypothesizes that IID events actually inhibit the occurrence of seizures. AIM 2 will determine the microphysiology of high frequency oscillations (HFOs) and their role in seizure initiation. Recent studies propose that HFOs (~ 80-250 Hz and > 250 Hz) in local field potentials are a hallmark of seizure initiation and might play a causal role. Yet, the relationship between these HFOs and single neuron activity in human epilepsy is unknown. Importantly, it is also unclear whether these HFOs are specific to epileptogenic neocortex or might have similar incidence rates even in healthy neocortex. We will thus compare the incidence and spatiotemporal properties of HFOs in epileptic neocortex in humans with focal epilepsy and in nonepileptic neocortex in human and nonhuman primates during sleep, rest and wake states. Finally, AIM 3 will determine the temporal evolution of neural synchrony at the level of single neurons during the preictal-to-ictal transition and during the seizure. Contrary to mainstream thought, recent animal models suggest that hyposynchrony, not hypersynchrony, initiates seizures.